Theoretical and numerical methods for predicting ship-wave impact generated sea spray

Mintu, Md. Shafiul Azam (2021) Theoretical and numerical methods for predicting ship-wave impact generated sea spray. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (4MB)

Abstract

Spray generated by ships traveling in cold oceans often leads to topside ice accretion, which can be dangerous to vessels. To develop a full methodology of goal based design for ice accretion there are two critical knowledge gaps, both of which are complex to close, and require new methods and techniques. One is a comparison of ice accretion rates for different structures in the same icing conditions. The second knowledge gap is validation data that compares predicted ice growth rates for all types of ship and offshore structures against observed values. Estimation of the spray flux is a first step in predicting icing accumulation. The amount of spray water, the duration of exposure to the spray, and the frequency at which the spray is generated are all important parameters in estimating the spray flux. Most existing spray flux formulae are based on field observations from small fishing vessels. They consider meteorological and oceanographic parameters but neglect the vessel behavior. Ship heave and pitch motions, together with ship speed and heading relative to the waves, determine the frequency of spray events. Thus the existing formulae are not generally applicable to different sizes and types of vessels. The current study develops simple methods to quantify spray properties in terms that can be applied to vessels of any size or type, which consequently addresses the first knowledge gap. Formulae to estimate water content and spray duration are derived based on principles of energy conservation and dimensional analysis. To estimate spray frequency considering ship motions, a theoretical model is proposed. The model inputs are restricted to ship’s principal particulars, operating conditions, and environmental conditions. Wave-induced motions are estimated using semi-empirical analytical expressions. A novel spray threshold is developed to separate deck wetness frequency from spray frequency. Spray flux estimates are validated against full-scale field measurements available in the open literature and reasonable agreement was obtained. The complex interaction between the structure and a multi-phase fluid, including spray are not fully understood. Limitations of field measurements and model experiments encourage the use of numerical simulation to understand the formation of such spray. In this study, full-scale simulation models of wave-generated sea spray are also developed by implementing a smooth particle hydrodynamics (SPH) method. A three-dimensional (3D) numerical wave tank equipped with a flap-type wave maker and a wave absorber is created to produce regular waves of various heights and steepness. A full-scale medium-size fishing vessel (MFV) is modeled to impact waves in head sea conditions at various forward speeds. Moving ship dynamics with three degree-of-freedom (3-DOF) in waves are resolved instead of mimicking a relative ship speed. The resultant spray water amount is measured using a numerical collection box and compared against field measurements and the theoretical model, where a reasonable agreement is found. The model is able to distinguish between green water and spray water. A multi-phase two-dimensional (2D) simulation is also performed that demonstrates the role of wind in the fragmentation of water sheets into droplets and their distributions over the deck. The simulation results indicate energy released from a surging ship significantly contributes to the generation of spray. An investigation was also performed to explore means to speed up the computationally intensive SPH simulations. A comparison with a traditional CPU (central processing unit) clusters with GPU (graphics processing unit) was performed where GPUs demonstrated faster executions. All the SPH simulations were run on GPUs.

Actions (login required)

View Item